Publication detail

GROUPS AND HYPERGROUPS OF ARTIFICIAL NEURONS

SMETANA, B. CHVALINA, J.

Original Title

GROUPS AND HYPERGROUPS OF ARTIFICIAL NEURONS

English Title

GROUPS AND HYPERGROUPS OF ARTIFICIAL NEURONS

Type

conference paper

Language

en

Original Abstract

When we study structure of the most used artificial neural network - multilayer perceptron and functionality of artificial neuron, there is possibility using several ways to describe function and neural network properties on the basis of known algebraic structures, vector spaces and graphs theory or properties of relations. Using certain analogy with relations between descriptions of differential equations certain quality there is developed access to new view point on these subjects. In this paper some concepts of description and modelling systems of neurons are investigated.

English abstract

When we study structure of the most used artificial neural network - multilayer perceptron and functionality of artificial neuron, there is possibility using several ways to describe function and neural network properties on the basis of known algebraic structures, vector spaces and graphs theory or properties of relations. Using certain analogy with relations between descriptions of differential equations certain quality there is developed access to new view point on these subjects. In this paper some concepts of description and modelling systems of neurons are investigated.

Keywords

Neural network, transposition hypergroups, linear ordinary differential operators, groups of neurons.

Released

05.02.2018

Publisher

Slovak University of Technology in Bratislava, Faculty of Mechanical Engineering

Location

Bratislava

ISBN

978-80-227-4765-3

Book

Proceedings, 17th Conference on Applied Mathematics – APLIMAT 2018

Edition

1

Edition number

1

Pages from

232

Pages to

243

Pages count

12

URL

Documents

BibTex


@inproceedings{BUT148455,
  author="Bedřich {Smetana} and Jan {Chvalina}",
  title="GROUPS AND HYPERGROUPS OF ARTIFICIAL NEURONS",
  annote="When we study structure of the most used artificial neural network - multilayer
perceptron and functionality of artificial neuron, there is possibility using several ways to
describe function and neural network properties on the basis of known algebraic structures,
vector spaces and graphs theory or properties of relations. Using certain analogy with relations
between descriptions of differential equations certain quality there is developed access to
new view point on these subjects. In this paper some concepts of description and modelling
systems of neurons are investigated.",
  address="Slovak University of Technology in Bratislava, Faculty of Mechanical Engineering",
  booktitle="Proceedings, 17th Conference on Applied Mathematics – APLIMAT 2018",
  chapter="148455",
  edition="1",
  howpublished="electronic, physical medium",
  institution="Slovak University of Technology in Bratislava, Faculty of Mechanical Engineering",
  year="2018",
  month="february",
  pages="232--243",
  publisher="Slovak University of Technology in Bratislava, Faculty of Mechanical Engineering",
  type="conference paper"
}